miR-181b-5p/SOCS2/JAK2/STAT5 axis facilitates the metastasis of hepatoblastoma

Abstract Introduction Hepatoblastoma (HB) is a malignant liver tumor predominantly found in children and tumor metastasis is one of the main causes of poor prognosis in affected patients. The precise molecular mechanisms responsible for HB metastasis remain incompletely understood. However, there is evidence suggesting a connection between the dysregulation of microRNAs (miRNAs) and the progression of tumor metastasis in HB. Methods The study utilized weighted gene co-expression network analysis (WGCNA) to analyze a miRNA microarray dataset of HB. The expression of miR-181b-5p in HB tissues and cells was detected using quantitative real-time PCR. The impact of miR-181b-5p on the metastatic capacity of HB was evaluated through scratch and Transwell assays. The effects of exogenously expressing miR-181b on the metastatic phenotypes of HB cells were evaluated in vivo. Furthermore, a luciferase reporter assay was performed to validate a potential target of miR-181b-5p in HB. Results We found that miR-181b-5p was highly expressed in HB tissues and HB cell lines. Overexpression of miR-181b enhanced scratch healing, cell migration, and invasion abilities in vitro, as well as enhancing HB lung metastasis potential in vivo. Dual-luciferase reporter assays showed that Suppressor Of Cytokine Signaling 2 (SOCS2) was a direct target of miR-181b. The overexpression of miR-181b resulted in the suppression of SOCS2 expression, subsequently activating the epithelial–mesenchymal transition and JAK2/STAT5 signaling pathways. The rescue experiment showed that SOCS2 overexpression attenuated the effects of miR-181b on HB cells. Conclusion Our study showed that miR-181b promotes HB metastasis by targeting SOCS2 and may be a potential therapeutic target for HB.


Introduction
Hepatoblastoma ( HB ) is a r ar e liv er tumor that primarily affects infants and young c hildr en. 1 Pulmonary metastases are frequentl y observ ed in HB cases, with ∼20% of patients pr esenting with metastases in the lungs.Unfortunately, the prognosis for c hildr en with HB and pulmonary metastases is gener all y poor. 2 The current standard treatment for HB pulmonary metastases involves a combination of chemotherapy and surgical resection, but < 50% of patients ac hie v e complete r emission after c hemothera py. 3 While sur gical r esection can effectiv el y cur e patients with primary liver tumors, those with pulmonary metastases miss out on this opportunity, significantly impacting their survival. 4Moreov er, c hemother a py r egimens for HB patients with pulmonary metastases ma y in volv e the administr ation of m ultiple a gents such as cisplatin, etoposide , and doxorubicin.T hese treatments necessitate long-term c hemother a py and can r esult in serious toxic side effects. 5Metastasis of HB is a major obstacle to effective clinical treatment due to the incomplete understanding of its k e y biological pathways and related functional genes.Further exploration and resear ch b y pediatric surgeons are needed to address this issue.
Weighted gene co-expression network analysis ( WGCNA ) is a bioinformatics tool used to study the structure and function of large-scale biomolecular networks .T he interactions between biomolecules form large-scale and complex biomolecular networks, and the analysis of the structure and function of these networks is crucial for understanding complex biological problems.
Identifying k e y nodes in the network system has always been an important issue in systems biology. 6Since microRNAs ( miRNAs ) ar e involv ed in the de v elopment of man y diseases , the disco very of disease-related molecules or regulatory mechanisms can be facilitated b y miRN A-mRN A netw orks. 7WGCN A can be used to identify signature genes or hub genes within modules to reveal gene interactions and regulatory mechanisms. 8A previous study used WGCN A to analyze RN A sequencing data and miRNAs expression profile data to identify a number of biomarkers for early tumor diagnosis. 9We collected miRNA expression profile data and predicted HB metastasis-related miRNA, and performed experimental validation and mechanistic analysis.
We identified miR-181b-5p, which exhibits a high correlation with HB metastasis, through the application of WGCNA.miR-181b-5p is a multifunctional miRNA that has been reported to participate in regulating the malignant biological behavior of various pediatric solid tumors.Abnormal expression of miR-181 family members was detected in 32 c hildr en with neur oblastoma, and ov er expr ession of miR-181b in vitro significantl y induced the gr owth and inv asion of neur oblastoma cells. 10Ewing's sarcoma is a common primary malignant bone tumor in c hildr en.A miRNA expr ession pr ofile anal ysis found that miR-181b is abnormall y expressed in the tumor and may serve as a potential prognostic marker for this inv asiv e pediatric bone tumor. 11These studies suggest that miR-181b plays an important r egulatory r ole in the de v elopment of pediatric tumors.Our bioinformatics analysis r e v ealed that miR-181b may play a potential role in HB metastasis.Ho w e v er, the exact mec hanism thr ough whic h miR-181b influences the pr ogr ession of HB r emains unclear, and the specific role of miR-181b in HB metastasis is also not well understood.This study aims to investigate the effect and molecular mechanism of miR-181b in HB metastasis, in order to provide a theoretical basis for the de v elopment of ne w str ategies for the pr e v ention and treatment of HB metastasis.

WGCNA
WGCN A w as conducted using the miRNA expr ession pr ofile data from the GSE153089 dataset.The tissue of this dataset included nontumor ous liv er tissue surr ounding HB ( n = 14 ) , primary HB tumor tissues of fetal subtype ( n = 11 ) and embryonal subtype ( n = 10 ) , and metastatic HB tumor tissue ( n = 9 ) .We used the WGCNA function pac ka ge in R softwar e to scr een for gene modules associated with disease phenotypes and further analyzed the relationship between these modules and HB tumor samples.The WGCNA function was utilized to r emov e unqualified genes and samples, resulting in the construction of a scale-free coexpression netw ork.The adjacenc y betw een genes w as calculated using the "soft" threshold po w er ( β) , which w as then transformed into a topological ov erla p matrix ( TOM ) .This matrix helped measure network connectivity and similarity.Based on the dissimilarity of TOM and a minimum gene number map ( n = 30 ) , genes with similar expression profiles were grouped into gene modules using av er a ge linka ge hier arc hical clustering and dynamic tree-cutting function detection.To further analyze the modules, the dissimilarity of module eigen genes was calculated.A cut line for the module dendr ogr am was c hosen, leading to the mer ging of some modules. 12We quantify associations of individual genes with our trait of interest ( tumor metastasis ) by defining gene significance ( GS ) as the correlation between the gene and the trait.The module membership ( MM ) was calculated as the correlation between the gene expr ession pr ofile and the module eigengene .T his measures the degree of membership of each gene within a module .T his allows us to quantify the similarity of all genes on the array to e v ery module.Differ entiall y expr essed genes ( DEGs ) between HB and normal liv er wer e scr eened fr om GSE153089 with the "edgeR" pac ka ge.Significantl y c hanged genes wer e selected with P v alue < 0.05 and log 2 | fold-change |≥1.

Tissue materials
We consecutiv el y collected the data and tissues of 12 patients with HB who underwent he patectom y at our institution between January 2021 and January 2023.The Ethics Committee on Biomedical Research, West China Hospital of Sichuan University appr ov ed this study ( 2019-1085 ) .With consent from patients or parents/legal guardians, tumor tissue and adjacent normal liver tis-sues were collected from patients with a confirmed diagnosis of HB.Collected tissues were stored in liquid nitrogen prior to molecular analysis.Informed, signed consent with regard to the collection and use of sample tissues in this study was obtained in adv ance fr om par ents/legal guardians of the patients .T he collection of samples was a ppr ov ed by the Ethics Committee of the West China Hospital of Sic huan Univ ersity.All information r egarding the human material was managed using anonymous numerical codes and the experiments were conducted following the Helsinki declaration.

Quantitati v e real-time PCR
The total RNA of eac h gr oup of cells was extr acted separ atel y according to the instructions of the Total RNA Extraction Kit ( TRIzol ) , and the quality of total RNA of each group was evaluated using a Nanodrop spectrophotometer.Reverse transcription was performed using 2 μg of RNA as template.After the cDNA w as obtained b y r e v erse tr anscription, the r eal-time PCR r eactions were performed in the SYBR Green One-Ste p Quantitati v e r ealtime PCR ( qRT-PCR ) Kit.The results wer e anal yzed by the 2 − CT method.

Western blot
To extract total protein from transfected cells in each group, lysis buffer was used.The protein samples were then incubated at 95 • C for 15 min to denature the proteins .T he denatured proteins were loaded onto a 10% SDS polyacrylamide gel for electr ophor esis and se paration.The se parated proteins were transferred onto a membr ane, whic h was then blocked with a 5% skimmed milk solution for 1 h at room temperature .T he membrane was subsequently incubated with the primary antibody overnight.After washing the membrane, it was incubated with the secondary antibody for 1 h.Finall y, the membr ane was exposed using an Electr ogener ated chemiluminescence system.

Scr a tch assay
Huh6 and HepG2 cells were diluted with serum-free dulbecco's modified eagle medium to a concentration of 2 × 10 4 /ml and inoculated in 6-well plates.When the cell confluence r eac hed 100%, a 200 μl pipette was used to make a scr atc h.After 48 h of continuous culture, the migration distance of cells was observed and recorded under a microscope.

Tr ans w ell assay
The transfected HB cells were counted and 10 5 cells were added to the upper Transwell chamber.Next, 500 μl of fetal bovine serum was added to the lower chamber.Then the chamber was incubated for 48 h before fixing and staining the cells with Giemsa stain.The number of cells that had passed through the membrane was counted by microscope.

Double-luciferase reporter assay
The miR-181b control plasmid and miR-181b ov er expr ession plasmid wer e co-tr ansfected with the 3'-UTR ( untr anslated r egion ) wild-type ( WT-SOCS2-3'-UTR ) and mutant ( MUT-SOCS2-3'-UTR ) SOCS2 vector into Huh6 cells.After 48 h of transfection, the supernatant was aspirated and discarded, and the cells were washed twice with PBS, follo w ed b y lysis with cell lysis buffer for 15 min, and the cell lysate was collected.Specifically, 100 μl of luciferase r eaction r ea gent was added to an opaque 96-w ell plate, follo w ed by the addition of 20 μl of cell lysate .T he plate was then shaken and the activity of the luciferase reporter gene was measured using a luminometer.Next, 100 μl of another r ea gent, Renilla lucifer ase r eaction r ea gent, was added to the same plate, and shaken, and the activity was measured.

Formation of lung metastases in vivo
The animal ethics committee of West China Hospital of Sichuan Univ ersity a ppr ov ed the animal studies ( Appr ov al Number: 20 230 106 004 ) .To explore the role of miR-181b in vivo , miR-181b-ov er expr essed and Gr een fluor escent pr otein ( GFP ) -labeled lentivirus ( L V -miR-181b ) and control lentivirus ( L V -NC ) wer e tr ansfected into Huh6 cells.A total of 10 nude mice were randomly divided into L V -NC group ( n = 5 ) and L V -miR-181b group ( n = 5 ) .Then the Huh6 cells ( 2 × 10 5 /200 μl ) were injected into the nude mice via tail vein.After 4 weeks, the mice were euthanized and the lungs were removed for imaging.An IVIS ® Lumina III in vivo imaging system was used to monitor for lung metastasis, imaged ex vivo .GFP luminescence was quantified as av er a ge r adiance using Living Image software version 4.0 and the region of interest ( ROI ) tool.

Sta tistical anal ysis
SPSS23.0 statistical software was applied for statistical analysis, t-test was applied to compare two groups, and one-way one-way analysis of variance ( ANOVA ) was used for comparison between m ultiple gr oups; P < 0.05 indicates that the results of comparison between experimental gr oups ar e statisticall y significant.All data are shown as mean ± SD from more than three independent experiments.Data were analyzed by Graph Pad Prism Version 9.0 software.

WGCNA screening for miRNAs associated with HB metastasis
We conducted WGCNA on the micr oRNA expr ession pr ofile.When β = 7 was set as the soft threshold, the weighted gene coexpression network conformed to a scale-free network ( Fig. 1 A ) .We constructed a systematic clustering tree and topological overlap matrix for miRNAs, identifying a series of modules formed by highly interconnected miRNAs ( Fig. 1 B ) .We found that the yellow module ( r = 0.33, P = 0.03 ) was closely related to HB metastasis ( Fig. 1 C ) , so we extracted the miRNAs from the y ello w molecular module for subsequent analysis.We selected miRNAs with MM > 0.8 and GS > 0.2 in the y ello w module as candidate k e y miRNAs ( Fig. 1 D ) .To further refine our investigation, we intersected the differ entiall y expr essed miRNAs in GSE153089 with the candidate miRNAs obtained from WGCNA, resulting in a selection of 22 miRNAs.Notably, miR-181b-5p exhibited the highest MM value of 0.97 ) .In the context of identifying core miRNA, MM can be used to prioritize miRNAs that have a strong correlation with a specific module of interest, so we focused on miR-181b-5p ( Table 1 ) .

miR-181b promotes the migr a tion and in v asion of HB cells
We examined the expression of miR-181b in 12 HB tissues and adjacent normal liver tissues .T he qR T-PCR sho w ed that miR-181b was significantl y upr egulated in HB tissue ( Fig. 2 A ) .Using human liver cell L02 as a control group, qRT-PCR revealed that miR-181b expression was higher in HB cells than in control cells ( Fig. 2 B ) .To further investigate the function of miR-181b in HB cell migration and inv asion, scr atc h and Tr ans well assa ys were performed.T he scr atc h assay sho w ed that upr egulation of miR-181b significantl y accelerated wound closure in HepG2 and HuH6 cells, while miR-181b knockdown decreased the wound closure capacity ( Fig. 2 C ) .Consistentl y, Tr ans well assa ys demonstrated that ov er expr ession of miR-181b significantly promoted the migration and invasion of HB cells, wher eas downr egulation of miR-181b significantl y inhibited the migratory and invasive ability of HB cells ( Fig. 2 D ) .Since miR-181b enhanced migration and invasion of HB cells, we aimed to explore whether miR-181b could promote HB lung metastasis in vivo .We generated stable clones of GFP-labeled Huh6 cells either with miR-181b ov er expr essed lentivirus ( L V -miR-181b ) or control lentivirus only ( L V -NC ) .Then these cells were injected into nude mice via tail vein and fluorescence signals in the ex vivo lungs were measured after 4 weeks.We found that lungs from the L V -miR-181b group had a higher fluorescence signal than those from the L V -NC group.These data suggest that miR-181b enhanced Huh6 cell metastasis in vivo ( Fig. 2 E ) .

miR-181b promotes epithelial-mesenchymal transition in HB cells
Tumor cells under going epithelial-mesenc hymal tr ansition ( EMT ) ar e mor e pr one to inv asion and metastasis.We investigated the relationship between miR-181b and EMT in HB cells.Western blot demonstrated that overexpression of miR-181b decreased the expression of e pithelial mark er E-cadherin but increased the expr ession le v el of mesenc hymal markers N-cadherin, vimentin, and snail in the HB cells .Con v ersel y, miR-181b inhibitor yielded the opposite effect on these EMT marker ( Fig. 3 ) .These in vitro experiments suggest that miR-181b promotes EMT in HB cells.

SOCS2 is a direct target of miR-181b-5p
To explore the underlying mechanism of miR-181b-5p, we conducted bioinformatics, and the results from the web server Starbase sho w ed that miR-181b complementarily binds to the 3'-UTR of SOCS2 ( Fig. 4 A ) .SOCS2 is downregulated in various human tumors and is associated with tumor metastasis and pr ogr ession. 13Ho w e v er, ther e ar e no r eports on the expr ession of SOCS2 in HB, nor ar e ther e an y r eports on its tar geting r elationship with miR-181b.To verify whether SOCS2 was a direct target of miR-181b-5p, we performed a dual-luciferase reporter assay on Huh6 cells .T he results sho w ed that miR-181b mimic specifically inhibited the luciferase activity of the wild-type ( WT ) SOCS2-3'UTR reporter plasmid in Huh6 cells, while it had no significant effect on the luciferase activity of the mutant ( MUT ) SOCS2-3'UTR reporter ( Fig. 4 B ) .Next, we used Western blot and qRT-PCR to investigate the effect of miR-181b on endogenous SOCS2 expression.After transfection with miR-181b-mimic in HB cells, the expression of SOCS2 r educed significantl y, at both mRNA and protein level.Conv ersel y, miR-181b-inhibitor tr ansfection led to the upregulation of SOCS2 expression ( Fig. 4 C,D ) .These results indicate that miR-181b has the potential to dir ectl y tar get and suppr ess the expr ession of SOCS2 in HB.

Inhibition of SOCS2 expression accounts for miR-181b-5p function in HB
To further verify whether the effects of miR-181b-5p on HB were mediated b y SOCS2, w e conducted r escue experiments.Ov erexpression of miR-181b significantly enhanced HB cell migration, while co-transfection with SOCS2 overexpression plasmid ( only the Coding sequence of SOCS2 was expressed, without the miRNA-181b targeting site in the 3 UTR ) r e v ersed this effect ( Fig. 5 A ) .Mor eov er, tr answell assay indicated that additional SOCS2 abrogated the promotion of migration and invasion of HB cells induced by miR-181b mimic ( Fig. 5 B ) .These results indicate that miR-181b-5p might promote HB cell migration and invasion through SOCS2.

miR-181b-5p enhanced the activ a tion of EMT via the JAK2/ST A T5 pathway
It has been reported that SOCS2 is a negative regulator of JAK2/ST A T5 signaling pathwa ys , which pla y a critical role in cell EMT. 14 , 15To better understand the relationship between miR-181b, SOCS2, and downstream signaling pathwa ys , we conducted a series of assa ys .Western blot results sho w ed that ov er expr ession of miR-181b enhanced the phosphorylation le v els of JAK2 and ST A T5, wher eas miR-181b inhibitor decr eased the expr ession of p-JAK2 and p-ST A T5 in HB cells ( Fig. 6 A ) .These results suggest that miR-181b activates the JAK2/ST A T5 pathway.We further explored the functional significance of JAK2/ST A T5 signaling in the pr o-metastasis r ole of miR-181b-5p in HB cells using the JAK2/ST A T5 signaling inhibitor Fedratinib ( TG101348 ) .Western blot results revealed that the stimulatory effect of miR-181b-5p on EMT and JAK2/ST A T5 acti vity was atten uated by Fedratinib ( Fig. 6 C ) , indicating that Fedratinib could reverse the effect of

The miR-181b-5p/SOCS2/JAK2/ST A T5 axis regulated HB metastasis
To investigate whether the pro-metastasis role of miR-181b-5p in HB cells was caused by the activation of the JAK2/ST A T5 pathw ay, w e emplo y ed a J AK2 inhibitor, Fedratinib .Trans well assa ys sho w ed that Fedratinib inhibited miR-181b-induced increases in HB cell migration and invasion ( Fig. 7 A ) .In addition, overexpression of SOCS2 partiall y r e v ersed the pr omotion of JAK2/ST A T5 and EMT activity by miR-181b-mimic in HB cells ( Fig. 7 B ) .The above results suggest that miR-181b-5p directly targets SOCS2, resulting in activation of JAK2/ST A T5 signaling in HB cells.

Discussion
The incidence of metastasis is a critical factor affecting the prognosis and survival of children with HB.T herefore , a comprehensive understanding of the underlying mechanisms of HB metastasis and identifying novel therapeutic targets is vital for improving the clinical outcomes of HB children.Our study sho w ed that dysregulation of the miR-181b/SOCS2/JAK2/ST A T5 axis is associated with metastasis of HB.Specifically, we found that upregulation of miR-181b in HB cells promotes tumor metastasis by inhibiting the expression of SOCS2, which leads to activation of JAK2/ST A T5 signaling and subsequent promotion of EMT related to in vasion.Con versely, restoration of SOCS2 expression or inhibition of JAK2/ST A T5 signaling can suppress tumor metastasis in HB.Taken together, our results suggest that the miR-181b/SOCS2/JAK2/ST A T5 axis may r epr esent a novel therapeutic target for the prevention and treatment of HB.
Identifying and c har acterizing the function of the miR-181b/SOCS2 axis in HB metastasis is significant.Firstly, as HB is typically asymptomatic in the early stages, understanding the role of the miR-181b/SOCS2 axis in this process may provide new insights into the mechanisms underlying HB progression and metastasis.Targeting this signal axis may be an effective strategy to treat metastatic disease in children with HB.Secondly, the miR-181b/SOCS2 axis is not known to regulate HB cell metastasis in pr e vious studies .T her efor e , our data ma y shed light on the molecular mechanisms that drive HB metastasis and identify potential nov el ther a peutic tar gets .T hirdly, identifying the function of miR-181b in HB may hav e br oader implications for cancer r esearc h and ther a py.Dysr egulation of miRNA expr ession is a hallmark of many cancers, and miRNAs may serve as potential diagnostic or prognostic biomarkers as well as therapeutic targets .T he miR-181 family is a highly conserved group of miRNAs that ar e incr easingl y important in biomedical r esearc h, and these miRNAs exhibit aberrant expression patterns in different solid tumors, where they can function as either tumor suppressors or promoters of cancer. 16One study found that the TGF β pathway regulates the transcription of miR-181b, which in turn promotes the pr ogr ess and c hemor esistance of liv er cancer. 17Neur oblastoma is a solid tumor that pr edominantl y affects c hildr en.Ther e is an upregulation of miR-181b in neuroblastoma, and this dysregulation   is significantly associated with the prognosis of children with neuroblastoma. 18Another important finding is that miR-181b was upregulated in MYCN-amplified neuroblastoma relative to the other tumor subtypes, while being decreased in retinoic acid-induced neur oblastoma. 19In contr ast to earlier findings, ho w e v er, some studies have shown that miR-181b has anti-cancer effects in other tumors.In glioma, miR-181b acts as a tumor suppressor by inhibiting tumor growth and invasion and inducing apoptosis, possibl y thr ough tar geting the NF-κB and EMT pathwa ys . 20Characterizing the role of miRNAs like miR-181b in HB may lead to the de v elopment of more effective and personalized therapies for a wide range of malignancies.
In our study, miR-181b has been shown to target SOCS2, a negativ e r egulator of the JAK2/ST A T5 signaling pathwa y.T he JAK2/ST A T5 signaling pathway is a crucial intracellular signaling pathwa y that pla ys a significant role in a variety of biological processes, including cell pr olifer ation, differ entiation, and surviv al. 21 Gr owing e vidence suggests that dysr egulation of the JAK/ST A T pathway is associated with tumor cell pr olifer ation, migr ation, and invasion. 22In the liver, the SOCS2/JAK2/STAT5 axis and the negativ e r egulatory r ole of SOCS2 ar e crucial for maintaining the normal physiological status of liver cells. 23Pr e vious studies hav e shown that high-glucose-induced JAK/ST A T activ ation upr egulates the expression of miR-181b in glomerular mesangial cells, establishing a link between JAK/ST A T signaling and miR-181b for the first time. 24We have identified miR-181b-5p as a k e y regulator of SOCS2 in HB metastasis.To the best of our knowledge, this is the first study to provide evidence of the involvement of the miR-181b-5p/SOCS2 axis in HB metastasis .T hese findings are highly novel and significant since they highlight a ne w r egulatory axis involved in HB metastasis.
EMT is a process by which epithelial cells acquire mesenchymal properties, allowing them to inv ade surr ounding tissues and migrate to distant organs, and activation of the JAK/ST A T Figur e 7. T he miR-181b-5p/SOCS2/JAK2/ST A T5 axis r egulated HB migr ation and inv asion.( A ) Fedr atinib decr eased the migr ation and inv asion of Huh6 cells in the miR-181b-5p ov er expr ession gr oup ( left: r epr esentativ e pictur es of Giemsa-stained migrated and invaded cells at 48 h time-point; right: the bar gr a phs show the number of migr ated and inv aded cells thr ough the membr ane ) .All quantifications wer e performed with thr ee independent r epeats, data ar e pr esented as means ± SD. * * * * P < 0.0001 b y one-w ay anal ysis of v ariance.( B ) Western blot anal ysis ( left ) and quantification ( right ) were performed to detect protein levels of J AK2, p-J AK2, ST A T5, p-ST A T5, E-cadherin, N-cadherin, vimentin, and snail in HB cells with indicated modifications.All quantifications were performed with three independent repeats, data are presented as means ± SD. * P < 0.05, * * P < 0.01, * * * * P < 0.0001 by one-way analysis of variance .ns , Not significant.
signaling pathway can induce cellular EMT and thus promote tumor metastasis. 25In our study, we have shown that the ov er expression of miR-181b leads to the downregulation of E-cadherin and upregulation of N-cadherin, snail, and vimentin, whic h ar e c har acteristic markers of EMT.Furthermore, we have demonstrated that the inhibition of miR-181b significantly suppresses EMT and tumor metastasis.Targeting miR-181b might be used to pr e v ent HB cells fr om under going EMT and acquiring inv asiv e properties, leading to reduced HB recurrence and metastasis.Ther e ar e se v er al limitations of this study.Firstl y, the study only focused on miR-181b and SOCS2 and did not consider other r egulatory factors involv ed in HB metastasis.Further r esearc h could investigate the involvement of other miRNAs and regulatory pathways in HB metastasis.Secondly, the sample size is relativ el y small, and more extensive studies with larger sample sizes could provide a more comprehensive understanding of the role of the miR-181b/SOCS2 axis in HB metastasis.

Conclusion
In summary, we have identified for the first time that miR-181b-5p dir ectl y tar gets SOCS2 in HB.In addition, we found that the biological effects of miR-181b-5p on HB metastasis were mediated through inhibition of SOCS2 and subsequent activation of the JAK2/ST A T5 signaling pathway.Our r esearc h suggests that miR-181b-5p may serve as a therapeutic target for HB patients.

Figure 1 .
Figure 1.Weighted gene co-expression network analysis to identify HB metastasis-related miRNAs.( A ) Analysis of network topology for various soft-thresholding po w ers .T he left panel shows the scale-free fit index ( y -axis ) as a function of the soft-thresholding po w er ( x -axis ) .The right panel displays the mean connectivity ( degree, y -axis ) as a function of the soft-thresholding po w er ( x -axis ) .( B ) Cluster dendr ogr ams r epr esenting gr oups of genes identified using weighted gene co-expression network analysis in HB datasets with the assigned module colors.( C ) Module-trait relationship of the significant modules correlating to HB, with correlation values to metastasis, embryonal, and fetal subtype phenotype.Each row corresponds to a module eigengene and each column corresponds to a trait.Each cell contains the corresponding correlation and P -value .T he figure is color-coded by correlation according to the color legend.( D ) A scatterplot of GS for metastasis versus MM in the y ello w module.

F igure 2 .
miR-181b-5p w as upr egulated in HB and pr omoted migr ation and inv asion.( A ) miR-181b-5p expr ession in human hepatoblastoma ( HB, n = 12 ) tissues and adjacent normal tissues ( ANT, n = 12 ) was determined b y qR T-PCR.The r elativ e expr ession le v els of miR-181b-5p in HB were higher than those in ANT.The violin plot displays the distribution of a continuous variable across different categories.* * * * P < 0.0001 by Student's t-test.( B ) Expr ession le v els of miR-181b-5p in HB cells ( Huh6 and HepG2 ) and LO2 cells .T hr ee r epetitions wer e carried out and data ar e pr esented as means ± SD. * * * * P < 0.0001 by one-way analysis of variance.( C ) Wound-healing assay analysis of HepG2 and Huh6 cells treated with either miR-181b-5p mimics or miR-181b-5p inhibitor versus scramble controls for 48 h ( left: re presentati ve brightfield pictures of cells at 0 or 48 h time-point; right: quantification of cells migrated into wound area ) .The width of the gap between two patches of cells was measur ed, scr atc h cov er ed r ate was calculated, and any significant differences between the control group and treatment group were identified.All quantifications were done with three inde pendent re peats.Data ar e pr esented as means ± SD. * P < 0.05, * * P < 0.01, * * * P < 0.001 by one-way analysis of variance .( D ) Trans well assa y analysis of HepG2 and Huh6 cells treated with either miR-181b-5p mimic or miR-181b-5p inhibitor versus scramble controls for 48 h ( left: re presentati ve pictures of Giemsa-stained migrated and invaded cells at 48 h time-point; right: bar graphs show the number of migrated and invaded cells through the membrane ) .The number of migrated and invaded cells was calculated and any significant differences between the control group and treatment gr oup wer e identified.All quantifications wer e done with thr ee independent r epeats.Data ar e pr esented as means ± SD. * * * * P < 0.0001 b y one-w ay analysis of variance.( E ) Re presentati ve fluorescence pictures of ex vivo lungs shown at completion of the 4-week study period ( left: GFP fluorescent images of lung metastases at the endpoint; right: the box plot shows the average bioluminescence signal measured in the lungs ) .The av er a ge r adiant efficiency measured in the lungs of L V -miR-181b group mice was higher than in lungs of L V -NC group mice ( n = 5 per group ) .Data are presented as means ± SD. * P < 0.05 by Student's t-test.

Figure 4 .
Figure 4. SOCS2 is a direct target of miR-181b-5p.( A ) The seed sequence of miR-181b-5p is complementary to the targeting site in the 3 UTR of SOCS2.( B ) Dual-luciferase assay validating SOCS2 as a target of miR-181b-5p.Luciferase reporter assay was performed in Huh6 cells using the synthetic miR-Control or miR-181-5p in combination with the WT or the MUT 3 UTR of the SOCS2 construct.All quantifications were done with three inde pendent re peats.Data ar e r epr esented as means ± SD. * * * * P < 0.001 b y one-w ay anal ysis of v ariance.( C ) Expr ession of SOCS2 mRNA in HB cells was assessed following alteration of miR-181b-5p expression using qRT-PCR.All quantifications were done with three independent repeats, and the expression of Glyceraldehyde -3-phosphate dehydrogenase ( GAPDH ) was used as a PCR internal reference.Data are represented as means ± SD. * P < 0.05, * * P < 0.01, * * * * P < 0.0001 b y one-w ay analysis of v ariance.( D ) Expr ession of SOCS2 protein in HB cells was analyzed by western blot following alteration of miR-181b-5p expression.Protein expression was quantified by band intensity and normalized to β-actin, values are expressed as mean ± SD ( n = 3 ) .* * P < 0.01, * * * P < 0.001, * * * * P < 0.0001 by one-way analysis of variance.

Figur e 5 .
Figur e 5. T he effects of miR-181b-5p on HB cell migration and invasion were partially reversed by overexpression of SOCS2.( A ) Wound healing assays were conducted to evaluate the motility of HB cells transfected with miR-181b-5p mimic alone or co-transfected with SOCS2.( Left: re presentati ve brightfield pictures of cells at 0 or 48 h time-point; right: quantification of cells migrated into wound area ) .The width of the gap between two patches of cells was measured and scratch cover rate was calculated.All quantifications were done with three independent repeats.Data are presented as means ± SD. * P < 0.05 by one-way analysis of variance .( B ) Trans well assa ys wer e conducted in HB cells tr ansfected with miR-181b-5p mimic alone or co-transfected with SOCS2.( Left: re presentati ve pictures of Giemsa-stained migrated and invaded cells at 48 h time-point; right: bar graphs show the number of migrated and invaded cells through the membrane ) .The number of migrated and invaded cells was calculated.All quantifications were done with three independent repeats.Data are represented as means ± SD. * * * * P < 0.0001 by one-way analysis of variance.

Figure 6 .
Figure 6.miR-181b-5p enhanced the activation of the JAK2/ST A T5 pathway and EMT in HB .( A ) W estern blot of J AK2, p-J AK2, ST A T5, and p-ST A T5 on HB cells transfected with miR-181b-5p mimic, inhibitor, or negative control.( B ) Quantitation of J AK2, p-J AK2, ST A T5, and p-ST A T5 pr otein using Ima geJ softwar e. Pr otein expr ession w as quantified b y band intensity and normalized to β-actin, v alues ar e expr essed as mean ± SD ( n = 3 ) .* * * * P < 0.0001 by one-way analysis of variance.( C ) Western blot of J AK2, p-J AK2, ST A T5, p-ST A T5, E-cadherin, N-cadherin, vimentin, and snail protein on HB cells ov er expr essing miR-181b or control vector and treated with Fedratinib.( D ) Quantitation of JAK2, p-JAK2, ST A T5, p-ST A T5, E-cadherin, N-cadherin, vimentin, and snail protein using ImageJ software.Protein expression was quantified by band intensity and normalized to β-actin, values are expressed as mean ± SD ( n = 3 ) .* * * * P < 0.0001 by one-way analysis of variance .ns , Not significant.